Statistically derived Phi-psi maps for each type of residue indicate substantial improvements in X-ray data over previous tabulations. Position effects in regular secondary regions show strong effects for some types of residues, especially proline, aromatic and polar groups. However, these maps do not coincide with calculated ones. We intend to investigate the origin of discrepancies in order to improve molecular calculations. NMR investigations of Substance P, a neuropeptide, indicate no dominant conformations in water, but in methanol it is highly ordered with a large number of 2-Dimensional NMR NOE cross-peaks. With these data, molecular models consistent with the experimental data have been developed. Subsequently additional data obtained have assisted in indicating which of the alternative conformations are most probable. Experimental studies of a 13 amino acid fragment of Ribonuclease A indicated it to be a stable helix at low temperatures. The existence of such a small relatively stable structure has been surprising. We have been looking in detail at possible molecular interactions to try to understand the remarkable stability of such a small peptide as a helix in water.